The linear model features were carefully studied in the cases of data perturbation and mean shift perturbation. Some important features were also proved mathematically. The results show that the mean shift perturbatio...The linear model features were carefully studied in the cases of data perturbation and mean shift perturbation. Some important features were also proved mathematically. The results show that the mean shift perturbation is equivalent to the data perturbation, that is, adding a parameter to an observation equation means that this set of data is deleted from the data set. The estimate of this parameter is its predicted residual in fact.展开更多
This paper proposes a robust method of parameter estimation and data classification for multiple-structural data based on the linear error in variable（EIV） model.The traditional EIV model fitting problem is analyzed...This paper proposes a robust method of parameter estimation and data classification for multiple-structural data based on the linear error in variable（EIV） model.The traditional EIV model fitting problem is analyzed and a robust growing algorithm is developed to extract the underlying linear structure of the observed data.Under the structural density assumption,the C-step technique borrowed from the Rousseeuw＇s robust MCD estimator is used to keep the algorithm robust and the mean-shift algorithm is adopted to ensure a good initialization.To eliminate the model ambiguities of the multiple-structural data,statistical hypotheses tests are used to refine the data classification and improve the accuracy of the model parameter estimation.Experiments show that the efficiency and robustness of the proposed algorithm.展开更多
文摘The linear model features were carefully studied in the cases of data perturbation and mean shift perturbation. Some important features were also proved mathematically. The results show that the mean shift perturbation is equivalent to the data perturbation, that is, adding a parameter to an observation equation means that this set of data is deleted from the data set. The estimate of this parameter is its predicted residual in fact.
基金supported by the National High Technology Research and Development Program of China (863 Program) (2007AA04Z227)
文摘This paper proposes a robust method of parameter estimation and data classification for multiple-structural data based on the linear error in variable（EIV） model.The traditional EIV model fitting problem is analyzed and a robust growing algorithm is developed to extract the underlying linear structure of the observed data.Under the structural density assumption,the C-step technique borrowed from the Rousseeuw＇s robust MCD estimator is used to keep the algorithm robust and the mean-shift algorithm is adopted to ensure a good initialization.To eliminate the model ambiguities of the multiple-structural data,statistical hypotheses tests are used to refine the data classification and improve the accuracy of the model parameter estimation.Experiments show that the efficiency and robustness of the proposed algorithm.
